America’s Expansion Has a Cash Cushion, Not Immunity

Executive Summary

The strongest case against an imminent U.S. recession is still the labor market’s stability. September unemployment was 4.2%, while the Sahm recession indicator stood at 0.00 percentage points, well below its 0.50-point trigger. Real GDP increased in the latest reported quarter, and the coincident recession model assigned only 0.62% to August. Those readings support an expansion baseline; none establishes that the next year is safe.

These are qualitative editorial assessments, not calibrated numerical forecasts. The August coincident-model reading and the year-end prediction-market contract answer different questions from these forward horizons.

The less obvious signal is liquidity. Money-market funds held $8.441 trillion in total financial assets at the end of Q2, up 12.8% from a year earlier. That is a substantial pool of liquid investments. But its size says little about who owns the assets or whether cash will finance consumption, capital spending or purchases of risky securities. Treating it as automatic “dry powder” would turn a useful balance-sheet observation into an unsupported investment thesis.

Labor Stability Is the First Line of Defense

Unemployment rose from 4.1% in July and August to 4.2% in September. A one-tenth-point increase merits attention, but it is not the sustained acceleration that typically makes recession risk self-reinforcing. The Sahm measure compares the three-month average unemployment rate with the lowest three-month average over the preceding 12 months. September’s zero reading indicates no gap under that definition.

The distinction between a level and a change matters. Unemployment is higher than the exceptionally low readings of 2022–23, yet recession warnings depend importantly on the speed of deterioration. A stable 4.2% rate can coexist with modest growth; an accelerating rate can weaken spending and hiring expectations even before GDP turns negative. The Sahm indicator is a contemporaneous warning rule, not an advance guarantee or an official recession declaration.

Recession Risk Scorecard

Output Is Expanding, but the Latest Quarter Is Not Today

Q2 real GDP was above Q1 by about 0.55%, equivalent to 2.22% annualized. The underlying level increased from $24,274.383 billion to $24,408.011 billion. This is evidence of continued production, not merely confidence or a market price. It is also backward-looking: the latest available GDP reference period ends in June.

The August recession-probability model provides a more recent, though still lagged, cross-check. Its rise from 0.24% to 0.62% is 0.38 percentage points—not evidence that recession has become likely. The model combines employment, industrial production, real personal income excluding transfers, and real manufacturing and trade sales. Its strength is breadth; its weakness for this task is that a smoothed historical-state estimate should not be mistaken for a prediction of next year’s outcome.

A Cash Cushion Can Support Resilience Without Fueling a Rally

Money-market assets increased by about $960 billion between Q2 2025 and Q2 2026. This can reflect a preference for liquid, income-producing assets, institutional treasury management or changes in the broader financial system. It does not reveal a single motive. High balances are consistent with both caution and resilience, which is why the indicator receives a contextual rather than directional verdict.

For investors, the practical question is where liquidity sits relative to obligations. A portfolio with planned withdrawals benefits from a liquid reserve; a highly leveraged company can remain vulnerable even when aggregate fund assets are large. Cash earns a place in risk management because it reduces forced-selling pressure, not because a national total tells investors when to buy equities.

Attention and Betting Prices Are Cross-Checks, Not Verdicts

Recession searches reached 2,038 on October 6. The increase against roughly a month earlier sits alongside declines against roughly a week and a year earlier. That mixed pattern does not support a simple claim that anxiety is accelerating. News attention can change without a corresponding change in employment or output, and the estimated volume measures a single keyword in one locale.

The dated Polymarket comparison is similarly narrow. MacroMicro reported 7.50% on October 3 for “U.S. recession by end of 2026.” The contract’s rules allow a Yes outcome from either two consecutive negative annualized real-GDP quarters within its specified Q2 2025–Q4 2026 window, or an NBER recession announcement concerning 2025 or 2026 made by the Q4 2026 advance GDP release. This differs from the NBER’s broader economic assessment and extends resolution beyond December 31. The quote is a historical market-implied price, not this report’s forecast or a guarantee of accuracy.

Professional Commentary and Outlook

The defensible stance is conditional confidence. Stable unemployment, a non-triggered Sahm rule, positive Q2 output and a low August coincident probability collectively argue against declaring a recession on October 8. The stale BCIg checkpoint adds historical context but deserves little weight in an October timing decision. More cash in the financial system improves some balance sheets’ flexibility; it cannot neutralize a shock to real income or employment.

Healthcare and consumer staples may offer comparatively stable demand, but valuation and company-specific leverage still matter. Industrials and discretionary businesses offer a clearer test of the expansion scenario, while financially leveraged firms face greater downside if revenues weaken. These are scenario exposures, not individual-stock recommendations. PortfolioAI’s public system listings provide a separate place to review rules-based portfolio signals; this macro assessment should be a cross-check, not a substitute for a defined investment process.

Bottom line: expansion is better supported than recession, but the case rests on actual activity—not the expectation that cash balances will rescue markets.